Multimodal recognition based on face and ear

Research of ear recognition and its application is a new subject in the field of biometrics authentication. As the ear is located at the side face, it is reasonable to combine ear with face for multimodal recognition. In this paper, a Full-space Linear Discriminant Analysis (FSLDA) is applied for recognition with ear images, face images and the combined ear and face images. Experiments are performed on USTB ear image database and ORL face database. Recognition rates show that multimodal recognition using both ear and face results in improvement over either face biometric or ear biometric.